The major goals of this work are to examine the time course, nature and durability of attentional guidance in visual search. When looking for a target among a field of distractors, observers have the ability to guide their attention to likely targets on the basis of features such as color and orientation. We have preliminary evidence that the search rate for a target increases as a function of set size, suggesting that attentional guidance becomes more efficient over time. The first aim of this proposal is to determine whether guidance is fully active at the beginning of a trial, or whether it takes time to develop. The second aim of the proposal is to measure the number of guiding representations that an observer can maintain at once. We have pilot data showing that Os can maintain at least three representations within the same dimension (color), as long as none of the target colors are also used as distractors. The final aim of this proposal is to characterize the decay rate of guidance. Once guidance becomes active, how long does it remain active? Exploring the timecourse, nature, and durability of attentional selection in this manner should help inform and constrain models of visual attention and visual search. [unreadable] [unreadable]